Sepsis, newly defined as organ failure caused by systemic response to infection1, contributes to half of all in-hospital deaths in the US, and is also the number one overall cost to the US healthcare system.2,3 Although sepsis outcomes have improved over the last decade with improvement in standardized sepsis care, mortality rates remain high (10-35%)4. Sepsis treatment still consists of source control, antibiotics, and supportive care. Despite dozens of clinical trials for immune-modulating intervention, no treatments specific for sepsis have been successfully brought to market5. Two consensus papers have argued that the failure of clinical trials is due to the massive patient heterogeneity in the sepsis syndrome, and our lack of tools for accurate molecular profiling5,6. Current tools, mainly clinical severity scores such as APACHE and SOFA, and the blood lactate level, are not readouts of the underlying inflammation in sepsis, but rather a crude look at the global level of patient illness.
Several groups have hypothesized that transcriptomic (genome-wide expression) profiling of the immune system via analysis of the whole blood may be an effective way to stratify sepsis patients. Important insights from these studies include overexpression of neutrophils proteases, a collapse in adaptive immunity, and an overall profound immune dysregulation in sepsis7-12. Some immune profiling techniques have been validated prospectively to show outcomes differences13,14, but no clinical tools have yet been translated into clinical practice. Still, most of these studies have been deposited in public databases for further re-analysis and re-use.
Thus, new molecular profiling tools are needed, both for improved patient care and resource stratification, but also as research tools for better clinical trials in sepsis.